Beyond Jan Vishwas: Why India must screen new laws, not just clean old ones
Regulatory impact assessment is essential for India to examine the true economic, judicial, and social costs of proposed laws before they are enacted into policy
India's regulatory framework excels at cleaning up historical rules but lacks a mandatory, independent impact assessment process for new legislation, leading to hidden economic costs, excessive compliance burdens, and overburdened courts.
India's regulatory framework excels at cleaning up historical rules but lacks a mandatory, independent impact assessment process for new legislation, leading to hidden economic costs, excessive compliance burdens, and overburdened courts.
India's regulatory framework excels at cleaning up historical rules but lacks a mandatory, independent impact assessment process for new legislation, leading to hidden economic costs, excessive compliance burdens, and overburdened courts.
In December 2025, the Union government gave the Parliament an unusually candid answer. When asked whether the Pre-Legislative Consultation Policy of 2014 had been evaluated, the Ministry of Law and Justice said it had not. It kept no record of which ministries complied with it. And it had no proposal to make consultation or legislative impact assessment binding.
That answer describes the Indian regulatory state accurately. The system knows what its rules say. It does not reliably know what they cost, or who bears that cost.
A regulatory impact assessment (RIA) is the instrument built for that gap: a short published document that travels with a draft law or rule. It states the problem, asks whether regulation is the right answer at all, prices the realistic options for business, citizens and the exchequer, and sets out how the rule will be measured. Most Organisation for Economic Co-operation and Development (OECD) governments require one. India has no comprehensive, independently scrutinised framework requiring it.
Cleaning up old rules, without screening new ones
India has built an impressive apparatus for cleaning up historically bad regulation. More than 40,000 compliances have been removed since 2014. The Jan Vishwas Act of 2023 decriminalised 183 provisions across 42 central Acts. Jan Vishwas 2.0, passed in 2026, modified 784 provisions across 79 Acts. Every one of these is a cleanup. None is a filter.
There is no requirement that a ministry, before notifying a rule, publish what the rule will cost, who will bear that cost, what alternatives were considered, and what evidence supports the choice. The Pre-Legislative Consultation Policy is executive guidance, , and compliance is neither tracked nor enforced. Only 44 of 186 Bills introduced between 2014 and 2019 were published for comment, So the stock keeps being cleaned while the flow goes unexamined.
Two rules that were never priced
Take the 2012 retrospective tax amendment. It was used to raise demands of about ₹1.1 lakh crore on 17 entities. The state collected ₹8,100 crore, of which ₹7,900 crore came from a single company. It then lost two investment treaty arbitrations, and in 2021 Parliament repealed the provision and refunded the money. Nothing in the 2012 process required anyone to set the expected revenue against treaty exposure, litigation cost and the price of investor doubt. An impact assessment would have forced that arithmetic into the open before the amendment moved.
Take the Promotion and Regulation of Online Gaming Act, 2025. It banned all real money games, skill and chance alike, putting over 2,000 startups and about two lakh jobs at risk. The failure is that no document compared prohibition with licensing, age gating or deposit caps.
Who actually pays
The economic cost of unassessed regulation is easy to describe and hard to see. A prominent regulatory policy think-tank counts 69,233 compliances and 6,618 annual filings across 1,536 Acts and rules that an Indian business must track. Each one was defensible on its own. Nobody added them up, because no process requires anyone to.
Compliance cost behaves like a fixed cost, which makes it regressive. A large firm hires a compliance team and spreads the expense over a big turnover. A 25-person firm pays the same filing in owner hours. The Economic Survey 2024-25 made the consequence explicit, noting that Indian firms often choose to remain small in order to stay below the regulatory radar, which holds back formalisation, labour productivity and employment growth.
The social cost is the mirror image, and it is the part the deregulation debate usually misses. India has 644 factory inspectors for 3,21,578 factories, roughly one for every 500. A standard that cannot be enforced at that ratio is not a protection. It becomes a discretionary power, exercised against whoever is unlucky or unconnected. Workers get the paperwork and not the safety. An ex ante assessment asks the awkward question early: does the state have the capacity to deliver what this rule promises?
Courts as the impact assessor of last resort
A large part of the court docket is regulation coming back as a dispute. This is the predictable product of three drafting choices: vague standards, criminal liability for procedural default, and retrospective operation.
Section 138 of the Negotiable Instruments Act is the clearest illustration. A 1988 amendment made cheque dishonour a criminal offence, to shore up the credibility of a cheque. When the Supreme Court took up the question in 2021, 35.16 lakh of the 2.31 crore criminal cases then pending were Section 138 matters, roughly one in seven
Without an ex ante assessment, the courts become the impact assessor of last resort. They do the job late, on an incomplete record, and after the cost has already been borne.
The questions worth settling
None of this settles the design, and the design is where this idea will be won or lost. Five questions are worth arguing about in specifics: where does the obligation live? A statute, a Cabinet decision and an executive policy produce very different levels of compliance, as 2014 demonstrated. Who does the scrutiny? A unit inside the sponsoring ministry is assessing its own proposal, and an external body needs standing and staff. What triggers a full assessment, and measured against what baseline, given that India has never systematically mapped its existing stock of rules? Does it reach subordinate legislation, regulators' circulars and state rules, which is where most Indian regulation actually lives and binds? Finally, what follows if a ministry simply does not do it?
The deregulation drive of the past decade has shown that the Indian state can repeal at scale. The harder question is whether the state can learn to reject a proposed rule when the evidence shows that a better alternative exists. That is a question about institutions rather than intentions, and it is now ripe for a serious argument about particulars.
Ankit Shrivastava is a consultant, Centre for Law, Policy and Governance, NFPRC Foundation.
Kartikey Agarwal is team manager at Centre for Law, Policy and Governance, NFPRC Foundation.